Your Messaging Strategy Doesn't Have a Technology Problem. It Has an Org Chart Problem.

A new BCG × Meta study put hard numbers on rich messaging. The number that actually matters isn't the one on the cover.

Every few months a new report lands explaining why conversational AI or rich messaging is the next great channel. The latest, from BCG and Meta, is one of the better ones: 533 enterprises across ten markets, confirming something practitioners have felt for years — companies that run messaging across more than one use case see roughly 2x the customer lifetime value and meaningfully better acquisition costs than those running a single campaign. The leaders post double-digit gains in CLTV, CAC, and engagement. The data is real.

But the finding that decides everything is buried on page 18 — and almost everyone is going to skim past it.

BCG calls it the 10/20/70 principle. To make messaging work at scale, you should spend 10% of your effort on AI, 20% on data and tech infrastructure, and 70% on the operating model — who owns the channel, how teams coordinate, how success is measured. Seventy percent. The least glamorous line item is the one that decides whether the other 30% ever pays off.

And here's the part that should make any executive uncomfortable: when BCG asked French leaders how important the operating model was, they rated it 43%. The people doing this well say 70%. That gap isn't a knowledge gap. It's a worldview gap. Most companies still believe the tool creates the value. It doesn't. The tool is the cheap part.

I've spent the last several years building messaging systems in production — hundreds of live workflows across insurers, banks, and telcos, mostly on WhatsApp, mostly in markets that don't show up in glossy European reports. So let me say plainly what a sponsored consulting deck can only imply: the reason your messaging program is stuck has nothing to do with AI maturity, channel readiness, or consumer appetite. It's that no one actually owns the channel, and the org chart is designed to keep it that way.

The myth of the pilot

The standard failure pattern looks like progress. Marketing runs a WhatsApp campaign. The open rates are absurd compared to email — often 4 to 5x the click-through. Someone presents a slide. The pilot is declared a success. Then it stalls, because the next logical step — order updates, delivery notifications, authentication, customer service — belongs to operations, to risk, to IT. And those teams have their own roadmaps, their own KPIs, and zero incentive to inherit a marketing experiment.

The BCG data confirms this without quite naming it: French firms overwhelmingly start in sales and marketing because it's the familiar use case, and most never leave. They keep treating the highest-leverage channel they have as a promotional toy. Meanwhile the use cases that build genuine trust — the order confirmation, the fraud alert, the "your flight gate changed" message — sit unbuilt, because they cross departmental lines no one wants to redraw.

This is the trap. The value of messaging compounds across the customer lifecycle. But organizations are structured down functional silos. You cannot capture a horizontal value with a vertical org chart. The pilot doesn't fail because it was wrong. It fails because the company is built to make sure it can never become anything bigger than a pilot.

Why the laggard markets are actually the teachers

The BCG report frames France as behind, and India, Brazil, and Indonesia as the leaders to emulate. Fair on the data. But the reason those markets lead is the part worth stealing, and it isn't strategy.

It's necessity. In markets where email never achieved primacy, where app downloads are expensive on data plans, where the phone is the primary computer, companies didn't choose messaging as a sophisticated omnichannel play. They were forced onto it as the default interface, early, with no fallback. And because it was the main channel from day one rather than a marketing add-on, it never belonged to a single function. It belonged to the whole company by default. The org design followed the channel, instead of the channel being crammed into the existing org design.

That's the actual lesson, and it inverts the usual reading. The advanced markets aren't ahead because they had better technology or smarter strategists. They're ahead because their constraints forced the operating model the 70% describes, before anyone had to argue for it in a steering committee. Europe has the opposite problem: mature legacy channels, entrenched ownership, and the luxury of treating messaging as optional. Sophistication is the handicap here, not the advantage.

If you operate in or sell into emerging markets, this is not a story about catching up. You may already be sitting on the operating model that European enterprises are about to spend three years and several consultants trying to build.

What the 70% actually requires

Strip away the framework language and the operating-model work comes down to a few unglamorous decisions most companies avoid because they're political, not technical:

• One owner, real authority. Messaging needs a single accountable owner with a mandate that crosses marketing, service, operations, and risk — not a committee, not a center of excellence with no budget. If the channel reports into marketing, it will die as a marketing tool. That's not a culture problem; it's a reporting-line problem, and it's fixable in an afternoon if leadership actually wants to.
• Shared metrics, or the silos win. As long as marketing is measured on campaign conversion and service is measured on ticket deflection, they will optimize against each other on the same channel. The leaders measure messaging on lifecycle outcomes — retention, resolution, fraud reduction, cost-to-serve — across functions. Pick the business outcome first; the operational metrics are downstream.
• Non-marketing use cases first, not last. This is counterintuitive and it's the highest-leverage move in the entire playbook. The transactional and authentication messages — the ones with obvious, non-promotional value — are what earn the consent and the open rates that make every later marketing message land. Companies that lead with the order update and the security alert build a channel customers trust. Companies that lead with promotions train customers to ignore them. Build the boring messages first. The valuable ones depend on it.
• Partners as leverage, governed tightly. Most enterprises lean on several external partners to ship this — BSPs, AI vendors, CRM platforms. The BCG number is around four partners per implementation, with roles that overlap or aren't aligned. Translation: vendor sprawl with no one steering. Partners are how you move fast on capability you'd be slow and expensive to build alone — but only if someone internal owns the architecture and holds them to it. Outsource the building. Never outsource the ownership.

The uncomfortable conclusion

The convenient reading of the BCG report is that messaging plus AI is a big opportunity and you should invest. True, and safe, and it will lead most companies to do exactly the wrong thing: buy more tools.

The honest reading is harder. The technology is largely solved and increasingly commoditized — that's what the 10% is telling you. AI will keep getting better whether or not you do anything. The data layer is real work but it's known work. The thing that will actually determine whether you win is whether your leadership is willing to take a channel away from the function that currently owns it, give it a real owner with cross-functional authority, and measure it on outcomes nobody currently shares.

That's not a technology decision. It's an org-design decision, and it's a political one. Which is precisely why most companies will keep buying tools instead — because reorganizing accountability is harder than approving a software budget, and it makes enemies.

The companies that win the next few years won't be the ones with the best AI. Everyone will have good AI. They'll be the ones who looked at their own org chart, admitted it was the bottleneck, and had the nerve to redraw it. The 70% was never about technology. It was always about who's brave enough to own the thing.